A rapid detection method for surface targets based on the application of unmanned boats

Abstract

The invention discloses a rapid detection method for water surface targets based on the application of unmanned boats, and belongs to the cross technical field of digital image processing and control systems. The present invention obtains target candidate areas through objective analysis. Since there may be certain false alarms in the candidate areas, significant areas are obtained by means of saliency analysis, and the objective and saliency are combined to eliminate false alarms and obtain accurate target positions. The present invention does not contain specific target type information, so it has good universality. Compared with other existing target detection algorithms, it has greatly improved both in the detection effect of the target and in the speed of the method. It has important guiding significance for the automatic obstacle avoidance of unmanned vehicles.

Description

technical field [0001] The invention belongs to the cross-technical field of digital image processing and control systems, and more particularly relates to a rapid detection method for surface targets based on the application of unmanned boats. Background technique [0002] In recent years, the research and development of unmanned boats have attracted the attention of many maritime military powers, among which the American "Spartan" unmanned boat and the Israeli "Protector" unmanned boat are more representative. . At present, whether it is from the perspective of civilian or military use, my country's demand for unmanned boats is increasing day by day, which is particularly urgent in areas such as territorial waters patrolling, combating piracy and smuggling. In the autonomous navigation of the unmanned boat, the rapid detection of surface targets is the basis for the automatic obstacle avoidance of the unmanned boat. Here are a few commonly used target detection methods: ...

AI Technical Summary

Problems solved by technology

In the autonomous navigation of unmanned boats, there are many types of targets (such as cruise ships, sailboats, warships, buoys, floating objects, reefs, etc.), and the attitude and perspective changes of the targets are very large. Therefore, the current target detection Algorithms cannot adapt well to real natural scenes

In addition, since the unmanned boat is oriented to practical applications, the real-time requirements for the algorithm are relatively high, and the current DPM and R-CNN algorithms are too complex, and the real-time performance is difficult to meet.

[0013] To sum up, although there are many related algorithms in target detection, it is difficult to apply them to the automatic obstacle avoidance of unmanned ships due to the universality and complexity of the algorithms.

Images